Road construction anti-collision warning method based on visual recognition

Abstract

The application relates to the technical field of construction safety, and discloses a road construction anti-collision early warning method based on visual identification, which comprises the following steps: when a second device or a construction worker is photographed by a camera, the relative position between the second device or the construction worker and a current first device is recognized according to a current camera image, the relative position comprises a lateral distance and a longitudinal distance between the second device or the construction worker and the current first device; route information of the current first device and the second device or the construction worker in a sensor is acquired; the estimated route of the current first device and the second device or the construction worker is estimated according to the route information; whether the estimated route of the second device or the construction worker and the estimated route of the current first device coincide with each other in a dangerous area is judged; and an alarm is simultaneously sent to the current first device and the second device or the construction worker in the case of coincidence. The application helps to prevent the collision between construction devices and construction workers, and also helps to avoid the collision between construction devices, so that the possibility of safety accidents in a construction site is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of construction safety, for example to a road construction anti-collision warning method based on visual recognition. BACKGROUND

[0002] In the process of road construction, there are situations where construction personnel and construction machinery are simultaneously located in a construction site for cross operation. Construction machinery such as road rollers and other equipment is large, and the driver has a blind spot in the field of vision. While working, the construction personnel also have difficulty in keeping track of the movements of the construction machinery at all times. If the driver and the construction personnel are not careful, there is a risk of collision between the man and the machine. There is also a risk of collision between two construction machines due to overlapping routes. Especially in poorly lit conditions such as at night or in tunnels, the risk of collision between the man and the machine is greater, resulting in unavoidable personal safety and property damage and accidents. Therefore, a road construction anti-collision warning method based on visual recognition is needed to solve the above problems.

[0003] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present application, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0004] To have a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not a general review, nor is it intended to determine key / important components or delineate the scope of protection of these embodiments, but as a prelude to the detailed description that follows.

[0005] The present disclosure provides a road construction anti-collision warning method based on visual recognition, comprising: for all running equipment and construction personnel in the construction site, the equipment is equipped with sensors and cameras around the equipment; the construction personnel are provided with sensors; the method comprises:

[0006] When the camera captures the second equipment or the construction personnel, the relative position between the second equipment or the construction personnel and the current first equipment is identified according to the current camera image, and the relative position includes the lateral distance and the longitudinal distance between the second equipment or the construction personnel and the current first equipment;

[0007] Obtain the route information of the current first equipment and the second equipment or the construction personnel in the sensor;

[0008] Estimate the estimated route of the current first equipment and the second equipment or the construction personnel according to the route information;

[0009] Determine whether the estimated route of the second equipment or the construction personnel overlaps with the estimated route of the current first equipment in a dangerous area;

[0010] The alarm is simultaneously sent to the current first device and the second device or the construction personnel in the presence of coincidence.

[0011] In one embodiment, the running device is provided with a reference object with scales and directions and can be photographed by the camera. After the camera is installed, the camera is calibrated to obtain the camera intrinsic parameters. When the camera photographs the second device or the construction personnel, the relative position between the second device or the construction personnel and the current first device is recognized according to the camera image, including:

[0012] According to the camera intrinsic parameters and the reference object, the distortion of the photographed reference object image is corrected to generate a corresponding relationship between the corrected image pixel coordinates and the reference object scales and directions;

[0013] According to the camera intrinsic parameters and the corresponding relationship between the corrected image pixel coordinates and the reference object scales and directions, the current camera image is corrected;

[0014] The target image of the corrected current camera image is obtained, and the target image includes the second device or the construction personnel;

[0015] The relative position between the target image and the current first device is calculated according to the target image and the reference object.

[0016] In one embodiment, the relative position between the target image and the current first device is calculated according to the target image and the reference object, including:

[0017] The target image and the camera image are segmented according to a preset segmentation model to generate the edge of the target image;

[0018] The left and right endpoint pixels of the edge of the target image are recognized and recorded;

[0019] The closest endpoint pixel to the current first device is determined according to the reference object;

[0020] The horizontal distance and the vertical distance between the closest endpoint pixel and the current first device are calculated according to the reference object.

[0021] In one embodiment, the estimated route of the current first device and the second device or the construction personnel is estimated according to the route information, including:

[0022] The working route distribution and the period and the corresponding time of the current first device and the second device or the construction personnel are recognized according to the route information;

[0023] The working range of the current first device and the second device or the construction personnel is evaluated according to the working route distribution;

[0024] The approximate overlapping area of the working range of the current first device and the second device or the construction personnel in the corresponding time period is recorded, and the approximate overlapping area includes a completely overlapping area and a preset approximate dangerous area.

[0025] The period information of the approximately overlapping area is recorded, and the routes and arrival times of the current first device and the second device or the construction personnel are estimated based on the period information of the approximately overlapping area.

[0026] In one embodiment, estimating the estimated route of the current first device and the second device or the construction worker based on the route information further includes:

[0027] Acquire multiple frames of current camera images without approximately overlapping areas;

[0028] Segmenting a target image in each frame of the current camera image according to a preset segmentation algorithm, wherein the target image includes the second device or the construction worker;

[0029] Obtain image information of a set part in the target image;

[0030] The estimated route of the target image is determined based on the image information of the set part of the target image in multiple consecutive frames of current camera images.

[0031] In one embodiment, when the target image is a construction worker, the set parts include feet, legs, and upper body; and determining the estimated route of the target image based on image information of the set parts of the target image in multiple consecutive frames of current camera images includes:

[0032] Compare the set part with the preset walking image to identify the moving direction of the target image;

[0033] Obtaining the relative position between the set part and the current first device;

[0034] Calculating the moving speed based on the change in relative position between the set part and the current first device in multiple consecutive frames of the current camera image;

[0035] Evaluate the estimated routes of construction crews based on their direction of travel and speed of movement.

[0036] In one embodiment, when the target image is a second device, the set part includes the endpoint pixel of the target image closest to the current first device; the estimated route of the target image is determined based on the image information of the set part of the target image in multiple consecutive frames of current camera images, including: evaluating the travel direction and moving speed of the second device based on the change in the relative position between the set part and the current first device in multiple consecutive frames of current camera images.

[0037] In one embodiment, the dangerous area includes a current first device area, a first dangerous area, and a second dangerous area, the first dangerous area is close to the current first device area, and the second dangerous area is outside the first dangerous area; the alarm is simultaneously sent to the current first device and the second device or the construction personnel in the case of overlap, including:

[0038] In the case of overlap with the second dangerous area, the alarm sends a flat alarm;

[0039] In the case of overlap with the first dangerous area, the alarm sends a variable alarm and a voice alarm to remind the position of the target about to collide.

[0040] The road construction anti-collision warning method based on visual recognition provided by the embodiments of the present disclosure can achieve the following technical effects:

[0041] When the camera captures the second device or the construction personnel, it indicates that the second device or the construction personnel is close to the current first device, at this time, after evaluating the routes of the current first device, the second device or the construction personnel, if there is overlap in the dangerous area, the alarm is sent to the current first device and the second device or the construction personnel to remind. In this way, it helps to prevent the collision of construction devices and construction personnel, and also helps to avoid the collision between construction devices, thereby reducing the possibility of safety accidents on the construction site, improving the safety of the construction site, and avoiding personal safety and property loss and accidents.

[0042] The general description above and the following description below are exemplary and explanatory only and are not intended to be limiting of the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0043] One or more embodiments are illustrated by way of example in the drawings and are not intended to be limiting of the embodiments. Identical reference numbers in the figures indicate similar elements and identical type designations indicate similar elements or steps throughout the various figures. The drawings are not intended to be to scale. In the drawings:

[0044] Figure 1 is a schematic diagram of a road construction anti-collision warning method based on visual recognition provided by the embodiments of the present disclosure. DETAILED DESCRIPTION

[0045] In order to enable a person skilled in the art to more fully understand the features and technical contents of the embodiments of the present disclosure, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the drawings, and the drawings are only used for reference and do not limit the embodiments of the present disclosure. In the following technical description, in order to facilitate explanation, a plurality of details are provided to provide a full understanding of the disclosed embodiments. However, one or more embodiments can still be implemented without these details. In other cases, well-known structures and devices can be simplified to show.

[0046] The terms "first", "second", and the like in the description and claims of the present disclosure and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion.

[0047] Unless otherwise specified, the term "a plurality of" means two or more.

[0048] In the present disclosure, the character " / " represents a "or" relationship between the objects before and after it. For example, A / B means: A or B.

[0049] The term "and / or" is a description of the association relationship between objects, which means that there can be three relationships. For example, A and / or B means: A or B, or, A and B, the three relationships.

[0050] The term "corresponding" can refer to an association or binding relationship, A corresponding to B means that there is an association or binding relationship between A and B.

[0051] It should be noted that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0052] In order to collect relevant data, sensors are installed on all operating equipment and construction personnel at the construction site, and cameras are installed around the equipment; the construction personnel are provided with sensors. The sensors here include sensors for detecting the moving position.

[0053] In combination with Figure 1 Based on the above collected relevant data, the road construction anti-collision warning method based on visual recognition provided by the present disclosure includes:

[0054] S101, when the camera shoots the second equipment or the construction personnel, the relative position between the second equipment or the construction personnel and the current first equipment is recognized according to the current camera image, and the relative position includes the lateral distance and the longitudinal distance between the second equipment or the construction personnel and the current first equipment.

[0055] Here, the lateral direction refers to the width direction of the current first equipment, and the longitudinal direction refers to the length direction of the current first equipment. The lateral direction and the longitudinal direction are perpendicular in a plane.

[0056] S102, acquiring the route information of the current first equipment and the second equipment or the construction personnel in the sensor.

[0057] S103, estimating an estimated route of the current first device and the second device or the construction personnel according to the route information.

[0058] S104, judging whether the estimated route of the second device or the construction personnel coincides with the estimated route of the current first device on the dangerous area.

[0059] S105, issuing an alarm to the current first device and the second device or the construction personnel simultaneously in the case of coincidence.

[0060] When the camera captures the second device or the construction personnel, it indicates that the second device or the construction personnel is close to the current first device. At this time, after the route of the current first device, the second device or the construction personnel is evaluated, if there is coincidence in the dangerous area, an alarm is issued to the current first device and the second device or the construction personnel to remind. In this way, it helps to prevent the collision of construction devices and construction personnel, and also helps to avoid the collision between construction devices, thereby reducing the possibility of safety accidents in the construction site, improving the safety of the construction site, and avoiding personal safety and property loss and accidents.

[0061] Optionally, considering that the image will be distorted when the camera shoots, the running device is provided with a reference object with scales and directions and can be shot by the camera, the camera is calibrated after the camera is installed, and the camera intrinsic parameter is obtained; when the camera shoots the second device or the construction personnel, the relative position between the second device or the construction personnel and the current first device is recognized according to the camera image, including: correcting the distortion of the shot reference object image according to the camera intrinsic parameter and the reference object, generating the corresponding relationship between the corrected image pixel coordinates and the reference object scales and directions; correcting the current camera image according to the camera intrinsic parameter and the corresponding relationship between the corrected image pixel coordinates and the reference object scales and directions; obtaining a target image of the corrected current camera image, the target image including the second device or the construction personnel; calculating the relative position between the target image and the current first device according to the target image and the reference object.

[0062] By calibrating the intrinsic parameter of the camera, the distortion of the reference object image is corrected through the scales and directions of the reference object, and then the distortion of the whole image is corrected according to the related parameters, so that the actual shape of the target image can be restored, which is convenient for subsequent accurate calculation of the horizontal distance and the vertical distance in the relative position, and helps to improve the accuracy of the alarm.

[0063] Optionally, the relative position between the target image and the current first device is calculated according to the target image and the reference object, including: segmenting the target image and the camera image according to a preset segmentation model to generate edges of the target image; identifying left and right end point pixels of the edges of the target image and recording; determining the end point pixel closest to the current first device from the left and right end point pixels according to the reference object; and calculating a horizontal distance and a vertical distance between the closest end point pixel and the current first device according to the reference object.

[0064] The segmentation model used in the embodiment can be a computer vision semantic segmentation model. The computer vision semantic segmentation model can be trained using the following method: collecting a training image dataset X of the current camera image; using a LabelMe labeling platform to label edge labels Y for the training image X; using a pre-trained FCN (fully convolutional deep semantic segmentation model) containing a large-scale human and mechanical device semantic segmentation image dataset PASCAL VOC, taking the training image dataset X of the current camera image as input and the human and mechanical device edge labels Y as a prediction target; constructing a loss function for the computer vision semantic segmentation model; and based on stochastic gradient descent, iteratively updating the semantic segmentation model parameters end-to-end to obtain the trained computer vision semantic segmentation model.

[0065] The computer vision semantic segmentation model is used to segment the target image and the camera image to find the end point pixel closest to the current first device, i.e. the point where a collision is likely to occur, and the horizontal distance and the vertical distance between the point where a collision is likely to occur and the current first device are calculated as related parameters of the relative position, which helps to accurately estimate the overlapping area on the route and thus improve the accuracy of the alarm.

[0066] Optionally, the estimated route of the current first device and the second device or the construction personnel is estimated according to the route information, including: identifying the work route distribution and period and the corresponding time of the current first device and the second device or the construction personnel according to the route information; estimating the work range of the current first device and the second device or the construction personnel according to the work route distribution; recording the approximate overlapping area of the work range of the current first device and the second device or the construction personnel in the corresponding time period, the approximate overlapping area including a completely overlapping area and a preset approximate dangerous area; recording the period information of the approximate overlapping area, and estimating the route and arrival time of the current first device and the second device or the construction personnel according to the period information of the approximate overlapping area.

[0067] Considering that the working equipment and the construction personnel of the general construction site have their own specific construction area, the working range and period of the current first equipment and the second equipment or the construction personnel are identified and the related time is recorded, the overlapping part in the corresponding time period is regarded as the area that may form a collision, that is, the approximate overlapping area, and the route and the arrival time of the equipment and the personnel in the approximate overlapping area are mainly evaluated. In this way, the evaluation range is reduced to the approximate overlapping area, the amount of data processed is reduced, and the evaluation speed is improved.

[0068] Optionally, the estimated route of the current first equipment and the second equipment or the construction personnel according to the route information further includes: in the case where there is no approximate overlapping area, acquiring a plurality of continuous current camera images; segmenting a target image in each current camera image according to a preset segmentation algorithm, the target image including the second equipment or the construction personnel; acquiring image information of a set part in the target image; and determining an estimated route of the target image according to the image information of the set part in the target image in the plurality of continuous current camera images.

[0069] Considering that there may be special situations such as displacement of the equipment or the construction personnel, that is, there is no certain rule, the image information of the set part in the plurality of continuous current camera images is used to determine the movement of the second equipment and the current first equipment, and the estimated route is evaluated through the movement, so as to accurately calculate and evaluate the movement route of the current first equipment and the second equipment or the construction personnel and the area that may overlap.

[0070] It should be noted that when the movement route of the second equipment or the construction personnel is acquired, the movement of the current first equipment also needs to be considered. When the real lateral distance and the longitudinal distance of the second equipment or the construction personnel, including the speed, are calculated from the relative position change, the movement of the current first equipment needs to be eliminated.

[0071] Optionally, in the case where the target image is the construction personnel, the set part includes a foot, a leg and an upper body; the estimated route of the target image according to the image information of the set part in the target image in the plurality of continuous current camera images includes: comparing the set part with a preset walking image to identify a moving direction of the target image; acquiring a relative position between the set part and the current first equipment; calculating a moving speed according to a change amount of the relative position between the set part and the current first equipment in the plurality of continuous current camera images; and evaluating the estimated route of the construction personnel according to the moving direction and the moving speed.

[0072] Optionally, in the case of the target image being the second device, the setting part includes the end point pixel closest to the current first device; and the judging of the estimated route of the target image according to the image information of the setting part in the continuous multiple frames of current camera images includes: judging the moving direction and moving speed of the second device according to the change of relative position between the setting part and the current first device in the continuous multiple frames of current camera images.

[0073] In consideration of the special situation that the device or the construction personnel may be displaced, i.e. there is no certain rule, the moving direction and speed of the device are relatively stable, and the moving direction and speed of the personnel may change greatly, the personnel and the device are judged separately.

[0074] When the person walks, the distance between the two feet, the direction of the foot, the bending of the leg, the inclination of the upper body and the inclination direction can be used as typical features to judge the walking direction of the person, and the typical features of the setting part of the human body are used to judge the estimated route of the construction personnel in the next period of time in combination with the continuous multiple frames of current camera images. When the device moves, the end point pixel closest to the current first device, i.e. the point that may collide, is selected to judge.

[0075] In this way, the personnel movement and the device movement are calculated separately, and the respective typical parts are compared and evaluated, so that the accuracy of the whole judgment process is high.

[0076] When the camera captures the construction personnel or the second device, it indicates that the construction personnel or the second device is already close to the first device, so when the estimated route of the construction personnel or the second device is evaluated, only a rough estimated route in a short time is needed to be evaluated.

[0077] For example, for the construction personnel, although the moving direction and speed of the construction personnel may change greatly, there is still a walking habit. When the person walks to the target position, it is usually divided into walking in a straight line or walking in a curve. Therefore, based on the walking habit of the person, the estimated route of the person is estimated according to the actual route and the typical features. Specifically, a route corresponding to the current continuous multiple frames of current camera images is obtained, and the route is matched with a spline curve, i.e. whether the route is a curve, an approximate straight line, or multiple approximate straight lines or a curve is judged.

[0078] In the case of an approximate straight line, it indicates that the construction personnel is used to walking in a straight line, so the walking direction determined by the typical features of the last frame of image is taken as the walking direction, and the estimated route of the current construction personnel in the case of walking in a straight line is calculated according to the speed.

[0079] In the case of a multi-segment approximate straight line, it is shown that the current construction personnel are used to walking in a straight line but do not proceed in a straight line towards the target position, so the walking direction determined by the typical features of the last frame image is taken as the walking direction, and the evaluation route of the current construction personnel in the case of walking in a straight line is calculated according to the speed.

[0080] In the case of a curve, the target position of the construction personnel is by default in the left front or right front, i.e., not walking a backtracking route. A plurality of consecutive tangent lines of the curve are drawn from the starting point of the curve to the ending point of the curve, the angle change of the tangent lines of the curve is obtained, if the angle change of the adjacent two tangent lines of the curve before the ending point is getting smaller and smaller until 0 and the angle change is less than or equal to a set a value, the direction adjustment of the construction personnel walking towards the target position is basically completed, so the walking direction determined by the typical features of the last frame image is taken as the walking direction, and the evaluation route of the current construction personnel in the case of walking in a straight line is calculated according to the speed; if the angle change of the adjacent two tangent lines of the curve before the ending point is getting smaller and smaller but the angle change is greater than the set a value or if the angle change of the adjacent two tangent lines of the curve before the ending point is getting larger, the direction deviation of the construction personnel walking towards the target position is large, so it is shown that the construction personnel are still adjusting the route, and according to the angle change, the starting point route direction is taken as the initial direction, the direction perpendicular to the starting point route direction is taken as the final direction, and the actual route is fitted to estimate the curve, and the estimated curve as a whole or one segment thereof can be taken as the evaluation route.

[0081] For the second device, since the advancing direction and speed of the device are relatively stable, the route can be evaluated according to the straight line and the direction obtained from the last frame image.

[0082] Optionally, the dangerous area includes a current first device area, a first dangerous area, and a second dangerous area, the first dangerous area is close to the current first device area, and the second dangerous area is outside the first dangerous area; in the case of overlapping, an alarm is issued to the current first device and the second device or the construction personnel at the same time, including: in the case of overlapping with the second dangerous area, the alarm emits a flat alarm; in the case of overlapping with the first dangerous area, the alarm emits a variable alarm and emits a voice alarm to remind the position of the target about to collide.

[0083] When the first device is in the second danger zone, it indicates that the distance between the first device and the second device is relatively close, and the collision will occur without changing the moving direction, but there is still a large reaction time to avoid, so the warning is a flat warning as a reminder. When passing through the second danger zone to the first danger zone, it indicates that the distance between the first device and the second device is close, and immediate action is required, which is relatively dangerous, so a variable warning is selected and a voice warning is issued to remind the driver and the second device or the construction personnel that may collide to take immediate action.

[0084] The above description and drawings suffice to fully illustrate the embodiments of the present disclosure to enable a person skilled in the art to practice them. Other embodiments can include structural, logical, electrical, process, and other changes. The embodiments only represent possible variations. Unless explicitly required, individual components and functions are optional, and the order of operations can be changed. Some embodiments can include or replace parts and features of other embodiments. Also, the words used in this application are only used to describe the embodiments and not to limit the claims. As used in the description of the embodiments and the claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms as well. Similarly, as used in this application, the term "and / or" refers to any and all possible combinations of one or more associated listed items. In addition, when used in this application, the term "comprise" and its variants "comprises" and / or "comprising" and the like refer to the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Without more limitations, the element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the process, method, or device that includes the stated element. In this document, each embodiment focuses on the differences from other embodiments, and the same or similar parts between embodiments can be referred to each other. For the method, product, etc. disclosed by the embodiments, if it corresponds to the method part disclosed by the embodiments, the relevant part can be referred to the description of the method part.

[0085] Those skilled in the art can clearly understand the technical solutions of the present application from the description of the embodiments disclosed in the present application. The present application can be implemented by means of software functional modules and combined logic, and the present application can also be implemented by means of hardware. Based on the technical concepts of the present application, those skilled in the art can implement the present application by means of software and / or hardware using the technical means disclosed in the present application, and the technical solutions disclosed in the present application can be implemented by means of a combination of software and hardware. Based on the technical concepts of the present application, those skilled in the art can implement the present application by means of software and / or hardware using the technical means disclosed in the present application, and the technical solutions disclosed in the present application can be implemented by means of a combination of software and hardware.

[0086] The flowcharts and block diagrams in the drawings show the possible implementation architecture, function and operation of the system, method and computer program product according to the embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code containing one or more executable instructions for implementing a specified logic function. In some alternative implementations, the functions noted in the blocks can also occur in an order different from that noted in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and sometimes they can also be executed in reverse order, depending on the functions involved. In the description corresponding to the flowcharts and block diagrams in the drawings, the operations or steps corresponding to different blocks can also occur in an order different from that disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and sometimes they can also be executed in reverse order, depending on the functions involved. Each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A road construction anti-collision warning method based on visual recognition, comprising: For all operating equipment and construction personnel at the construction site, sensors are installed on the equipment and cameras are installed around the equipment; The construction personnel are provided with sensors; the method includes: When the camera captures the second device or the construction worker, identifying the relative position between the second device or the construction worker and the current first device based on the current camera image, the relative position including the horizontal distance and the vertical distance between the second device or the construction worker and the current first device; Obtaining current route information of the first device and the second device or the construction personnel from the sensor; Estimating the estimated routes of the current first device and the second device or the construction personnel based on the route information; including: Identify the current working route distribution and cycle of the first device and the second device or the construction personnel and the corresponding time according to the route information; Evaluate the current working scope of the first and second equipment or construction personnel based on the distribution of work routes; Recording the approximate overlapping area of ​​the working ranges of the first device and the second device or the construction personnel within the corresponding time period, wherein the approximate overlapping area includes a completely overlapping area and a preset approximate danger zone; Recording period information of the approximately overlapping area, and estimating the routes and arrival times of the current first device and the second device or the construction personnel based on the period information of the approximately overlapping area; Acquire multiple frames of current camera images without approximately overlapping areas; Segmenting a target image in each frame of the current camera image according to a preset segmentation algorithm, wherein the target image includes the second device or the construction worker; Obtain image information of a set part in the target image; Determining an estimated route of the target image based on image information of a set portion of the target image in multiple consecutive frames of current camera images; Determine whether the estimated route of the second device or the construction worker overlaps with the dangerous area on the current estimated route of the first device; In the event of overlap, an alarm is simultaneously issued to the current first device and the second device or the construction personnel.

2. The road construction anti-collision warning method based on visual recognition according to claim 1 is characterized in that: The operating device is provided with a reference object with scale and direction and can be photographed by a camera. After the camera is installed, the camera is calibrated to obtain the camera internal parameters. When the camera photographs a second device or a construction worker, the relative position between the second device or the construction worker and the current first device is identified based on the camera image, including: Correcting the distortion of the captured reference object image based on the camera's intrinsic parameters and the reference object, and generating a correspondence between the pixel coordinates of the corrected image and the scale and direction of the reference object; Correct the current camera image based on the correspondence between the camera intrinsic parameters and the pixel coordinates of the corrected image and the scale and direction of the reference object; Acquire a target image of the corrected current camera image, where the target image includes the second device or the construction worker; The relative position between the target image and the current first device is calculated according to the target image and the reference object.

3. The method according to claim 2, characterized in that Calculating a relative position between the target image and the current first device according to the target image and the reference object includes: Segment the target image and the camera image according to a preset segmentation model to generate the edge of the target image; Identify the left and right endpoint pixels of the edge of the target image and record them; Determine the endpoint pixel closest to the current first device among the left and right endpoint pixels according to the reference object; The horizontal distance and the vertical distance between the nearest endpoint pixel and the current first device are calculated according to the reference object.

4. The method according to claim 1, wherein In the case where the target image is a construction worker, the set parts include feet, legs, and upper body; and determining the estimated route of the target image based on image information of the set parts of the target image in multiple consecutive frames of current camera images includes: Compare the set part with the preset walking image to identify the moving direction of the target image; Obtaining the relative position between the set part and the current first device; Calculating the moving speed based on the change in relative position between the set part and the current first device in multiple consecutive frames of the current camera image; Evaluate the estimated routes of construction crews based on their direction of travel and speed of movement.

5. The method according to claim 1, wherein In the case where the target image is a second device, the set part includes the endpoint pixel of the target image closest to the current first device; judging the estimated route of the target image based on the image information of the set part of the target image in multiple consecutive frames of current camera images includes: evaluating the travel direction and moving speed of the second device based on the change in the relative position between the set part and the current first device in multiple consecutive frames of current camera images.

6. The method according to claim 1, characterized in that The danger zone includes the area where the first device is currently located, a first danger zone, and a second danger zone, wherein the first danger zone is close to the area where the first device is currently located, and the second danger zone is outside the first danger zone; when there is overlap, an alarm is issued to the first device and the second device or the construction personnel at the same time, including: In case of overlap with the second danger zone, the alarm will send out a level-adjusted alarm; In the case of overlap with the first danger zone, the alarm emits a modified tone alarm and a voice alarm to remind the location of the target that is about to collide.

Images